Machine Learning

Linear regression explained from scratch: what the line means, how least squares chooses it, a worked example you can follow by hand, and the assumptions to check.

Feature scaling for beginners: normalisation vs standardisation vs robust scaling, which algorithms need it, and the leakage bug almost everyone hits.

Overfitting and underfitting are opposite failures. Diagnose which one you have from two numbers, then fix it with six proven, practical techniques.

Why you must split training and test data, how big each set should be, when to add a validation set, and the four leakage mistakes that inflate your scores.

A machine learning model is a function whose numbers are learned from data. Understand parameters, features, labels, training and inference in plain English.

Loss functions are how a machine-learning model measures its own mistakes. Every model learns by making the number a loss

Hinge loss is the objective behind support vector machines, and it thinks about classification differently from probability-based losses. Instead of

Binary cross entropy is the loss you reach for whenever a model must answer a yes-or-no question: spam or not,

The cross entropy loss is the workhorse objective for classification, the way squared error is for regression. It measures how

Huber loss is the clever compromise between the two great regression losses. It behaves like the smooth, fast-training mean squared